This project aims to investigate the status of associative recognition memory in global amnesia in the context of a dual-process model that postulates two separate bases for recognition performance, recollection and familiarity. Four studies will be conducted to examine this issue. The goal of Experiment 1 is to determine whether associative recognition for unrelated word pairs, as compared to single word recognition, is disproportionately impaired in amnesia. Experiment 2 compares associative memory for stimuli in which there is a pre-existing representation of an association (i.e., compound words) to stimuli in which there is not (i.e., unrelated words pairs). Experiment 3 examines whether associative recognition in amnesia is better preserved under the former conditions. Finally, Experiment 4 aims to specify the nature of the familiarity process contributing to amnesics? associative recognition (i.e., conceptual or perceptual) for compound stimuli.